Changes between Version 5 and Version 6 of BluePrintDecisionMakingTechnical


Ignore:
Timestamp:
10/02/09 12:50:23 (13 years ago)
Author:
Connie White
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • BluePrintDecisionMakingTechnical

    v5 v6  
    5656
    5757the difference between these and the regular calculation (scale 1) is that in scale 1 only the active voters are considered so the total voters is simply the total voters who actively voted on something (not everyone who could) because remember, they don't have to vote on every item, only those items (subgroup) they select.
     58
     599.3 Thurstone’s Law of Comparative Judgment
     60
     61
     62Thurstone’s Law of Comparative Judgment (TLCJ) measures a person’s preference for one item i over any another item j based on a single stimulus measuring the discriminal dispersion between the two on a psychological continuum.  This is conducted by using paired comparisons where every item i in a pair (i,j) in a set will be compared to every other item in a set producing a total of n(n-1)/2 comparisons (Thurstone, 1927). 
     63
     64Torgerson provides an example of this process using three matrices: matrix F (frequency), matrix P (probability) and matrix X (unit normal derivative).  The first matrix F, is a frequency count of item selected in any given pair by every individual in the group.  The table presented next shows a frequency count of a user group where N = 5.  The items along the 1st row, A, B, C, and D are preferred over the corresponding items in the 1st column of an M x M matrix, where zeros are placed along the diagonal.  Given the number of users, N = 5, each corresponding pair should equal N, i.e. (i,j) + (j,i) = 5: 
     65
     66        A       B       C       D
     67A       ----    1       2       1
     68B       4       ----    3       4
     69C       3       2       ----    5
     70D       4       1       1       ----
     71Table x.1 Matrix F Frequency Count of User Input
     72
     73These frequencies are changed to probabilities as such for matrix P where each each (i,j) + (j,i) = 100%:
     74
     75        A       B       C       D
     76A       ----    .20     .40     .20
     77B       .80     ----    .60     .80
     78C       .60     .40     ----    1.00
     79D       .80     .20     .0      ----
     80        2.2     .80     1.0     2.0
     81Table x.2 Matrix P Frequencies Converted to Probabilities
     82
     83The final matrix, X, is where these percentages are replaced by their unit normal deviates to cumulative proportions.
     84
     85        A       B       C       D
     86A       ----    -.85    -.26    -.85
     87B       .84     ----    .25     .84
     88C       .25     -.26    ----    0
     89D       .84     -.85    0       ----
     90        1.93    -1.96   -.01    -.01
     91Table x.3 Matrix X Cumulative Normal Distribution Function
     92
     93Values greater than 50% will be positive in this transformation of matrix P to X and values less than 50% will be negative.  Any values of 100% or 0% in matrix P are given a value of 0 in matrix X because the x values corresponding from the unit normal deviate table are unboundedly large (Torgerson, 1958).
     94
     95Thurstone’s scale is an extension based off of Thorndike’s work which provides added insight into ranked lists of items.  Given A is preferred over B 75% of the time and B is preferred over C 85% of the time, “how much greater than the distance AB is the distance BC?  Thorndike solved this problem by assuming that the difference in distances is proportional to the difference in the unit normal deviates corresponding to the two proportions” (Torgerson, 1958, p. 155).   
     96